NVIDIA Launches Nemotron 3 Nano Omni: Open Multimodal Model for 9x More Efficient AI Agents

NVIDIA released the Nemotron 3 Nano Omni model on April 28, 2026, an open omni-modal reasoning model unifying vision, audio, and language for up to 9x more efficient AI agent workflows including computer use, document intelligence, and audio-video reasoning. Available via Hugging Face, OpenRouter, build.nvidia.com, and 25+ partner platforms, the model is open and customizable for deployment anywhere. Written by Kari Briski, NVIDIA VP of generative AI and inference software.